The more subtle the differences, the greater the cohort numbers required to establish statistical relevance.
With too many variables and with really minute differences, it's highly likely that no definitive statistical conclusion will ever be reached.
That's my feeling for what we're dealing with here. There are a LOT of variables associated with shooters, their techniques, rifle quality, barrel characteristics, load differences, on and on. With an almost infinite amount of clutter to deal with, there will probably too much glare to see the light.
DF
